American Statistical Association
Propensity score methods such as nearest-neighbor matching and inverse-probability treatment weighting (IPTW) have been proposed as a means to recover balance between groups of interest and mimic the sample that would have been observed in a randomized trial. When estimating propensity scores, missing covariate data is a major issue that is commonly overlooked, leading to suboptimal matching or improper IPTW. Multiple imputation (MI) is a natural procedure to handle missing data in this context. However, there are open issues regarding the implementation of MI for propensity score analysis. In this talk, we will (1) investigate two opposing proposed methods to combine the MI and propensity score analysis steps, and (2) address variance estimation of the IPTW estimators after MI.
|Date:||Wednesday, April 3, 2019|
|Time:||4:00 - 5:00 P.M.|
Memorial Sloan Kettering Cancer Center
Department of Epidemiology and Biostatistics
485 Lexington Avenue
(Between 46th & 47th Streets)
2nd Floor, Conference Room B
New York, New York
**Outside visitors please email firstname.lastname@example.org for building access. You must be on the security list to enter the floor.